A mock score tells you where you stand today. It does not tell you why, and it does not tell you what to do tomorrow. Two aspirants with the same score can have completely different problems — one attempts too few questions, the other attempts too many and bleeds marks to negatives. The analysis is where you find out which one you are.

A useful rule of thumb: plan to spend at least as long analysing a mock as you spent taking it. If that sounds like a lot, it is a sign the mock is being used as a test rather than as practice.

Before you analyse: take the mock properly

Analysis is only as good as the data. Take full mocks in one sitting, in a quiet room, ideally at the same time of day as your exam slot. Do not pause, do not look anything up, and do not "just check" an answer mid-section. A mock taken casually produces numbers that describe a casual attempt.

Step 1: Get the numbers straight

Start with four numbers for each section: questions attempted, correct answers, wrong answers, and score. From those, work out:

  • Accuracy — correct answers divided by questions attempted.
  • Marks lost to negatives — under the marking scheme used in recent CAT papers (+3 for a correct answer, −1 for a wrong multiple-choice answer, no penalty on non-MCQ "type in the answer" questions), every wrong MCQ costs you one mark outright, plus the three you did not earn.
  • Attempt rate — attempts as a share of the questions in the section.

Our CAT mock analyser calculates these for you and lets you keep a private log of every mock on your own device. Always confirm the current marking scheme on the official CAT website (opens in a new tab) — it is set by the conducting institute each year.

Step 2: Sort every question into four buckets

This is the most valuable part of the process. Go through the paper question by question and put each one in a bucket:

  1. Attempted and correct. Check whether you got there efficiently. A correct answer that took six minutes is still a problem.
  2. Attempted and wrong. These need a diagnosis (step 3).
  3. Skipped, but doable. Questions you could solve comfortably once you sit with them, unrushed. This bucket is your biggest opportunity — these are marks you were capable of but did not reach.
  4. Skipped, and rightly so. Genuinely hard or long questions. Leaving them was the correct call.

Count each bucket per section. Most aspirants are surprised by how large bucket 3 is. It usually points to question selection — not ability — as the main lever.

Step 3: Diagnose every wrong answer

For each wrong answer, write down one cause:

  • Concept gap — you did not know the idea or formula needed.
  • Calculation or silly error — right method, wrong arithmetic, or a sign slip.
  • Misread the question — you solved a different question from the one asked.
  • Time-pressure guess — you answered without really solving it.

The fix is different for each. Concept gaps need study. Calculation errors need slower, cleaner working and practice of mental arithmetic. Misreads need a habit of re-reading the final line before marking. Time-pressure guesses need a decision rule (see below).

Step 4: Audit your time

Most mock platforms show time spent per question. Look for the questions where you spent well over the section's average time per question. Ask two things: did the time produce a correct answer, and could you have known earlier that the question was long?

The aim is not to answer every question faster. It is to recognise earlier which questions to leave.

Step 5: Re-solve before reading solutions

Before you open the official solution, re-attempt every question in buckets 2 and 3 without a timer. You learn far more from finding the method yourself than from reading someone else's. Then read the solution and note any approach that was faster than yours.

A note on guessing

Under a +3/−1 scheme with four options, a completely random guess has an expected value of zero: a one-in-four chance of +3 and a three-in-four chance of −1. Eliminate even one option confidently and the expected value turns positive. That is the logic behind a simple rule: do not guess blind, but an educated guess after eliminating options is reasonable. Your own accuracy data should tell you how reliable your eliminations really are.

Step 6: Turn it into three actions

End every analysis by writing down three specific actions for the coming week — for example, "revise remainders and divisibility", "in DILR, read all sets for two minutes before choosing", "re-read the question stem before marking any RC answer". Keep these in an error log alongside your scores so you can see whether last week's actions worked.

If you prefer a spreadsheet, the mock test tracker does the score and accuracy maths for you as you log each mock.

How many mocks should you take?

There is no magic number. What matters is that every mock is followed by a proper analysis. Taking mocks back to back without analysing them mostly measures the same weaknesses repeatedly.

This article explains a general method. Exam patterns and marking schemes can change between years, so always check the latest notification on the official website before your exam.

Written by Akash Gautam (LinkedIn profile, opens in a new tab). Spotted an error? Tell us.